Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- llama2
|
| 4 |
+
- llama2-70b
|
| 5 |
+
- rkllm
|
| 6 |
+
- rockchip
|
| 7 |
+
- rk3588
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
# Llama 2 Chat 70B for RK3588
|
| 11 |
+
This is a conversion from https://huggingface.co/meta-llama/Llama-2-70b-chat-hf to the RKLLM format for Rockchip devices.
|
| 12 |
+
This runs on the NPU from the RK3588.
|
| 13 |
+
|
| 14 |
+
# But wait... will this run on my RK3588?
|
| 15 |
+
No. But I found interesting to see what happens if I converted it.
|
| 16 |
+
Let's hope Microsoft never knows that I was using their SSDs as swap because they don't allow more than 32 GB RAM for student subscription :P
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
And this is before finishing, it will probably go to 600 GBs of RAM.
|
| 21 |
+
|
| 22 |
+
But hey! You can always try yourself getting a 512 GB SSD, a 32 GB of RAM SBC, have some patience and see if it loads. Good luck with that!
|
| 23 |
+
|
| 24 |
+
# Main repo
|
| 25 |
+
See this for my full collection of converted LLMs for the RK3588's NPU:
|
| 26 |
+
|
| 27 |
+
https://huggingface.co/Pelochus/ezrkllm-collection
|
| 28 |
+
|
| 29 |
+
# License
|
| 30 |
+
Same as the original LLM:
|
| 31 |
+
|
| 32 |
+
https://huggingface.co/meta-llama/Llama-2-70b-chat-hf/blob/main/LICENSE.txt
|